Journal of Chromatographic Science 2015;53:1265– 1273
doi:10.1093/chromsci/bmu224 Advance Access publication February 5, 2015
Study on Pharmacokinetics of Three Preparations from Levistolide A by LC – MS-MS
Wen-Qing He1, Wei-Sheng Lv1†, Yu Zhang1†, Zhao Qu1, Ren-Rong Wei1, Lei Zhang1, Chang-Hui Liu1, Xin-Xin Zhou2, Wei-Rong Li1,
Xiao-Tao Huang1‡ and Qi Wang1*
1
Institute of Clinical Pharmacology, Guangzhou University of Chinese Medicine, 12 Jichang Road, Guangzhou 510405, China, and
2
School of Continuing Education, Guangzhou University of Chinese Medicine, Guangzhou, China
*Author to whom correspondence should be addressed. Email: wqitcm@qq.com
†
These authors contributed equally to this work and should be considered as co-first authors.
‡
This author contributed equally to this work.
Received 10 September 2014; revised 6 December 2014
A rapid sensitive analytical method was established and validated
to investigate levistolide A in rat plasma by liquid chromatography –
tandem mass spectrometry operated in the positive ion mode.
Levistilide A (LA) and internal standard (IS) andrographolide (AD),
mixed with the plasma sample, were separated on a reversed
phase SpursilTM C18 5 mm column. The precursor/product transi-
tions (m/z) were 398.5/381.3 for LA and (m/z) 368.0/351.1 for AD.
The calibration curve was linear over the range from 5 to 1,250 ng/mL
for oral administration and 10 – 4,000 for intravenous administration
with a correlation coefficient (r) 0.9993. The lower limit of quanti-
fication was 5 ng/mL for LA in plasma. The inter- and intra-day accu-
racy and precision were less than + 15% of the relative standard
deviation. In this study, the developed method is successfully applied
to the comparative pharmacokinetic study of LA in rats after oral ad-
ministration of LA alone, Rhizoma Chuanxiong, and Danggui-Shaoyao-
San along with the bioavailability study of LA in rats. Our study shows
that low bioavailability (7.5%) is observed after oral administration of
LA. Traditional formula compatibility of Danggui-Shaoyao-San could
significantly enhance LA bioavailability compared with LA alone
and Rhizoma Chuanxiong.
Introduction
Danggui-Shaoyao-San (DSS), consisting Paeonia lactiflora Pall,
Angelica sinensis, Rhizoma Chuanxiong, Poria cocos,
Atractylodes macrocephala and Alismatis, is a common formula
in traditional Chinese medicine. It not only has blood-activating
and stasis-eliminating pharmacological effects for the treatment
of gynecological disorders, but also has an antidepressant-like ef-
fect for the treatment of insomnia that is a symptom and predic-
tor of depression (1, 2). In addition, it possesses the ability for
improvement of the cognitive functions and the neuroprotective
dysfunctions, leading to an effective improvement of Alzheimer’s
patients in Asian countries (3 –5). Recently, it is proved to ame-
liorate deterioration of cognition in SAMP8, especially in female
animals. Increasing estradiol (E2), NO and glycine might contrib-
ute to the cognitive improvement effect of DSS in female SAMP8
(6). Meanwhile, DSS also contributes to improve myocardial
ischemia (7).
Rhizoma Chuanxiong, which belongs to Umbelliferae family,
as a significant constituent part of DSS, is the dried rhizome of
# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
Article
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
Ligusticum chuanxiong Hort. It is widely used in the prescrip-
tions of traditional Chinese medicines (8). It has been confirmed
to have antioxidative effects (9). In addition, it is significantly ef-
fective in the treatment of headache and vertigo as well as vital-
izing blood circulation to treat cardiovascular diseases, such as
angina pectoris, ischemic stroke, migraines, menstrual disorders,
amenorrhea, dysmenorrhea, abdominal pain with mass forma-
tion, headaches and rheumatic arthralgia (10, 11). Besides, it
was reported for dispersal of tissue stasis, removal of chronic in-
flammation and facilitation of tissue perfusion (12). The elemen-
tary biologically active components of Rhizoma Chuanxiong
are organic acids, alkaloids and phthalides (9, 13). Moreover,
extracts of lobed kudzuvine root and Rhizoma Chuanxiong
were investigated for their in vitro of beta-amyloid (Ab1-42)-
aggregation-and acetylcholinesterase-inhibitory activities (10).
Levistilide A (LA), polymerized by two molecular ligustilide,
has been demonstrated to be the active components in
Rhizoma Chuanxiong (14). It is responsible for antimycobacte-
rials as well as curing cardiovascular diseases (15, 16). In addi-
tion, LA is available for sensitizing multidrug resistance
medicine cells to various chemotherapy drugs (17). It was also
proved that levistolide A and vinorelbine have a synergic antican-
cer effect on human non-small cell lung carcinoma (18).
To evaluate the determination of LA in plants or pharmaceutical
samples, several methods have been reported, which include chro-
matographic techniques, NMR spectroscopy and tandem mass
spectrometry(MS) (16) and high-performance liquid chromatogra-
phy with diode-array detection electrospray tandem mass spec-
trometry (HPLC – DAD) (3, 19). However, because of the poor
selectivity and more time consumption, it raises the necessity to es-
tablish a method with high sensitivity, specificity, reliability and sim-
plification to determine the pharmacokinetic parameters of LA.
Zuo et al. identified absorbed components and metabolites of
Rhizoma Chuanxiong decoction by ultraviolet (UV) detection,
tandem mass spectrometry (MS) and MS/MS (15). However, its
analysis condition was unsuited to biological samples. In addition,
there are no publications regarding the simultaneous determina-
tion of LA, Rhizoma Chuanxiong and DSS extract for the target of
LA and the application of the pharmacokinetics so far. Therefore, in
this study, a robust and selective HPLC coupled with mass spec-
trometry (LC –MS) method was used for the quantitation of LA
in rat plasma. It may be a turning point in the quantification of
LA as well as bioavailability after i.g. and i.v. administration.
Experimental
Instruments and reagents
The samples were analyzed by liquid chromatography – tandem
mass spectrometry (LC – MS-MS), which consists of an Agilent
1200 series liquid chromatographic system (containing a
G1311A Quart pump, a G1322A degasser, a G1313A automatic
sampler (ALS) and a G1316A thermostatted column compart-
ment; Agilent Technologies, USA), interfaced to a API4000 triple
quadruple mass spectrometer (Applied Biosystems/MDS-SCIEX,
Foster City, CA, USA) and equipped with electrospray ionization
(ESI). Data acquiring and processing were carried out on Analyst
1.5.1 software package (Applied Biosystems, MDS-SCIEX,
Toronto, Canada), which also controlled the LC–MS-MS system.
The separation was achieved on a SpursilTM C18 column (150 
4.6 mm, 5 mm; Waters Corp., Milford, MA, USA). Levistolide A
(.98% purity, LA) was obtained from Chengdu Herbpurify Co.,
Ltd (Chengdu, China). Andrographolide (.98% purity, AD) was
purchased from National Institutes for Food and Drug Control
and used as the internal standard (IS). Figure 1 shows the chem-
ical structures and fragmentation pattern of LA and AD.
HPLC-grade methanol was supplied by Merck (Darmstadt,
Germany). The purified water used for the LC – MS-MS analysis,
which was produced with a Millipore water purification system
(Millipore, Bedford, MA, USA). HPLC-grade ammonium acetate
was purchased at CNW Technologies GmbH (Germany). All
other chemicals and solvents were of analytical grade. Freshly ob-
tained drug-free rat plasma was collected from male
Sprague-Dawley (SD) rats (supplied by Medical Experimental
Animal Center of Guangzhou University of Chinese Medicine)
in our laboratory and reserved at 2808C until use.
Preparation of Rhizoma Chuanxiong extract
One kilogram dry powder of Rhizoma Chuanxiong ( purchased
from Xing Yuanchun Chinese Pharmacy, Guangzhou, China) was
Figure 1. Chemical structures and MS/MS spectra showing prominent precursor to
product ion transitions of LA (A) as well as IS AD (B).
1266 He et al.
immersed overnight in 10-fold volume of 95% ethanol. The mix-
ture was extracted under reflux twice each 2 h. After filtration,
the supernatants were desiccated in vacuo. Then the dried res-
idue was compounded in 0.3% sodium carboxy methylcellulose
(CMC-Na) suspension. Afterward, we obtained an oral solution of
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
Rhizoma Chuanxiong decoction with 2 mg/mL of LA in it.
Preparation of Danggui-Shaoyao-San extract
Totally 2,610 g dry powder of DSS including 900 g P. lactiflora,
270 g A. sinensis, 270 g Rhizoma Chuanxiong, 360 g P. cocos,
360 g A. macrocephala and 450 g Alismatis ( purchased from
Xing Yuanchun Chinese Pharmacy) was immersed overnight in
10-fold volume of 95% ethanol. The mixture was extracted
under reflux twice for 2 h. The supernatants were filtered before
evaporating to desiccation in vacuo. Then the dried residue was
also compounded in 0.3% CMC-Na suspension. Finally, an oral
solution of DSS decoction with 2 mg/mL of LA in it was
procured.
Preparation of calibration samples and quality
control samples
The primary stock solutions of LA (66.6 mg/mL) and the IS AD
(230 mg/mL) were prepared in methanol, respectively. The cali-
bration curves were made as follows: one is for oral administra-
tion, which was obtained by diluting the mixture of the stock
standard solutions with methanol at effective concentrations of
5, 10, 50, 125, 250, 500, 1,250 and 2,500 ng/mL. Another set of
the concentration of calibration curve line was as the same man-
ner to prepare the concentrations of 10, 25, 100, 200, 400, 1,000,
2,000 and 4,000 ng/mL for intravenous administration. All solu-
tions were stored at 48C and were brought to room temperature
prior to use. Calibration curve samples for measuring LA were
freshly prepared by spiking 10 mL working standard solution
with 100 mL drug-free rat plasma. Quality control (QC) samples
were separately prepared in the same manner as the calibration sam-
ples at low (10 and 25 ng/mL), medium (500 and 1,000 ng/mL) and
high (1,000 and 3,000 ng/mL) LA levels for i.g. and i.v., respec-
tively. The IS working solution (AD, 500 ng/mL) was diluted by
the primary stock solutions. Further processing of both the cali-
bration curve samples and QC samples were executed as the pro-
cedure mentioned in preparation of plasma sample section.
Animals
Thirty-two male SD rats with a weight of 250 + 10 g were ob-
tained from Laboratory Animal Center of Guangzhou University
of Traditional Chinese Medicine. All of the rats were sustained
under controlled air-conditioned conditions (temperature,
22 + 28C; relative humidity, 50 + 10%, commutative 12 h light –
dark cycles). The rodents were supplied free food and water in-
take and acclimatized to our animal room for 7 days. The animal
investigation was conducted according to the Regulations of the
Animal Ethics Committee of Guangzhou University of Traditional
Chinese Medicine.
All rats were fasted for 12 h and had free access to water prior
to dosing. Thirty-two male SD rats were randomly divided into
four groups with eight rats in each. Three groups were by oral
administration at the doses of 20 mg/kg of LA. The first group
was given an Rhizoma Chuanxiong solution containing LA at a
concentration of 2 mg/mL in 0.30% CMC-Na suspension. The
second group was given a DSS solution containing LA at a con-
centration of 2 mg/mL inside 0.30% CMC-Na suspension. The
third one was managed to acquire an LA standard solution with
a concentration of 2 mg/mL in 0.30% CMC-Na suspension.
Another one group was assigned to receive a dose of 2 mg/kg
LA standard solution intravenously, which was prepared in phys-
iologic saline. Serial blood samples (500 mL) were collected into
heparinized plastic tubes via the postorbital vein plex from each
rat at 0 h (before dosing) and subsequently at 0.083, 0.25, 0.5, 1,
1.5, 2, 4, 6, 8 and 12 h after oral dosing and at 0.033, 0.117, 0.25,
0.5, 0.75, 1, 2, 4, 6, 8 and 12 h after intravenous dosing. The blood
samples were rapidly centrifuged at 3000g for 10 min at 48C to
obtain plasma. These plasma samples were frozen at 2808C
until LC–MS-MS analysis.
Preparation of plasma sample
The solid-phase extraction (SPE) cartridges (ProElut C18
200 mg/3 mL 50/pkg; Dikma Technology, USA) were washed
with 3,000 mL of methanol followed by 3,000 mL of water. After
the frozen plasma samples (100 mL) were thawed out at room
temperature, 10 mL IS (500 ng/mL) and 790 mL water were
spiked with it. The resultant mixture were loaded to the SPE car-
tridges, then washed with 1,000 mL of water. Clean 1.5 mL
Eppendorf tubes were positioned under the SPE cartridges and
the objective compounds were eluted with 1,000 mL of metha-
nol. The eluent was evaporated to dryness by a gentle stream
of nitrogen at 378C. The residues were reconstituted in 100 mL
of the mobile phase (methanol). After vortexing for 10 min and
centrifuging for 10 min at 12,000g at 48C to remove the protein
precipitate, 50 mL supernatants were pipetted into glass auto-
sampler vials that contained inserts and analyzed by LC –MS-MS.
Liquid chromatographic conditions
The LC mobile phase was methanol – 1 mM ammonium acetate
(90 : 10, v/v) solution. Compounds were separated under iso-
cratic conditions at a flow rate of 0.25 mL/min. The chromato-
graphic run time for each injection was 5 min. The column
temperature was set to 358C using a column heater.
Mass spectrometric experiments
Typical sets of operation parameters used in this study were
under the following conditions: the probe temperature was
4008C, the nebulizer gas (GS1), turbo gas (GS2) and curtain gas
(CUR) were nitrogen and separately set to 35, 25 and 15 psi. The
ion spray voltage was set at 4,500 V. The sensitivity of detection
of both LA and IS in the positive ion mode was found to be higher
than that in the negative mode. So positive ESI mode was used,
and the ions were performed in the multiple reaction monitoring
(MRM) mode, in which we monitored the transition selected for
determination. The dwell time was at 200 ms. Optimal com-
pounds parameters including declustering potential (DP), en-
trance potential (EP), collision energy (CE) and collision exit
potential (CXP) were 40.4, 8.9, 11.2 and 15 V and 40.3, 7.2,
11.1 and 12 V for LA and IS, respectively. The corresponding pa-
rameters for each analyte are summarized in Table I.
Table I
Optimized Mass Parameters for LA and the Internal Standard AD
Analyte MRM (m/z) DP (V) EP (V) CE (V) CXP (V)
LA 398.40 ! 381.4 40.4 8.9 11.2 15.0
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
AD 368.20 ! 297.10 40.3 7.2 11.1 12.0
Method validation
The analytical method was validated to satisfy the US Food and
Drug Administration (FDA) bioanalytical method validation guid-
ance (20). Validation parameters were discussed in the following.
Matrix effect
The matrix effects in rat plasma, except for analytes, may cause
ion inhibition or enhancement of the signal. An assessment of
matrix effect was determined by comparing peak areas (A) of
the standard analytes in spiked blank plasma (five different sourc-
es) with the corresponding peak areas (B) obtained by injection
of analytes added in the mobile phase at appropriate concentra-
tion. The matrix effect for LA was investigated at QC of low, me-
dium and high concentrations, whereas the matrix effect over
the IS was determined at a single concentration of 500 ng/mL.
The peak area ratio of A/B (as a percentage) was used to measure
quantification for the matrix effects. It was identified that the ma-
trix effects did not have an obvious effect if the ratio is ,85% or
.115% (21).
Stability
The stability was tested with untreated QC samples at three con-
centration levels, i.e., low (10 and 25 ng/mL), medium (500 and
1,000 ng/mL) and high (1,000 and 3,000 ng/mL) LA levels for i.g.
and i.v., respectively, using five replicates at each concentration
and each storage condition. The long-term stability was assessed
for 30 days of storage at 2208C. The freeze – thaw stability was
investigated over three freeze – thaw cycles (2208C to room
temperature as one cycle). The post-preparative stability was de-
termined at room temperature for 24 h in the same manner. The
stability was expressed by the relative error (RE). Samples con-
formed to be stable if assay values exceed the acceptance criteria
of accuracy and precision (85–115%) (22).
Extraction recovery
The recovery of LA was tested at three QC levels (n ¼ 5), respec-
tively. Recoveries were calculated by comparing the mean peak
area of the extracted QC sample with that of the standard solution
containing the equivalent amount of analyses (23). The recovery of
LA was operated at low (10 ng/mL), medium (500 ng/mL) and
high (1,000 ng/mL) for oral administration and 25, 1,000 and
3,000 ng/mL for i.v. concentrations, while the IS was evaluated
at a single concentration of 500 ng/mL, which was used through-
out the analysis.
Calibration curve
The calibration curve was a line obtained by spiking pooled drug-
free plasma with an appropriate amount of working solution
to produce the calibration curve points whose concentration
was 5, 10, 50, 125, 250, 500, 1,250 and 2,500 ng/mL for oral ad-
ministration and 10, 25, 100, 200, 400, 1,000, 2,000 and 4,000 ng/
mL for intravenous administration. Besides, IS was 500 ng/mL.
LC– MS-MS determination of LA 1267
The correlation coefficient (r) of calibration curve was found to
be 0.9993 or better.
Lower limit of quantification
The lower limit of quantification (LLOQ) was determined as the
concentration that can be accurately measured, 10 times the
signal-to-noise (S/N) ratio.
Accuracy and precision
To determine the accuracy and precision of this method, three
QC concentration levels containing low (10 and 25 ng/mL), me-
dium (500 and 1,000 ng/mL) and high (1,000 and 3,000 ng/mL)
LA levels for i.g. and i.v., respectively, and they were done in blank
plasma for analysis. The measured value was expressed as the
mean of the five values. The accuracy was determined by
Figure 2. Representative LC–MS-MS chromatograms of blank plasma (A), LLOQ of LA and IS spiked with blank plasma (5 and 10 ng/mL, respectively) (B), blank plasma spiked with
1016 ng/mL of LA and 500 ng/mL of IS (C) and rat plasma at 1.5 h following a single oral administration of DSS at 20 mg/kg (D).
1268 He et al.
comparing the difference between the nominal value and mea-
sured value expressed as the RE. The precision was defined as
the coefficient of variation and expressed as relative standard
deviation (RSD) (24). For the evaluation of intra-day accuracy
and precision, five aliquots of each concentration were analyzed
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
in 1 day. For inter-day accuracy and precision, five replicates of
each concentration were evaluated on 3 days. The criteria for ac-
ceptability of accuracy and precision should be within +15%
(25).
Specificity and selectivity
The assay were assessed by comparing the chromatograms from
five different lots of drug-free rat plasma with those mixed with
respective standards to detect if there any interferences at the LC
peak region for LA and IS from endogenous plasma components.
All the plasma samples were processed and analyzed using the
same procedure and LC –MS conditions as mentioned.
Pharmacokinetic analysis
The relevant pharmacokinetic analysis was performed as follows:
such as methanol, ammonium acetate and formic acid along
with altered flow rates (in the range of 0.15– 0.3 mL/min) were
tested for complete chromatographic resolution of levistolide A
and IS. It was also concluded that the optimal mobile phase con-
sisted of methanol—1 mM ammonium acetate (90 : 10, v/v) at a

the maximum concentration (Cmax) and the time-to-maximum
concentration (Tmax) were determined by the experimental
data. Other pharmacokinetic parameters maintaining biological
half-life (t1/2) and total body clearance (Cl) were calculated by
using NONMEM Program version 1.1 (GloboMax Inc., Ellicott
City, MD, USA). The elimination rate constant (k) was estimated
from the terminal linear portion of the log plasma concentra-
tion–time curve, and the half-life (t1/2) of the drug was obtained
by 0.693/k. The area under the plasma concentration – time
curve (AUC) from time zero to the last measurable time point
(AUC0! t ¼ 8.0 h for i.v. administration, t ¼ 24.0 h for i.g. admin-
istration) and from time zero to infinity (AUC0! 1) were estimat-
ed using the trapezoidal rule-extrapolation method (26).
Results
Liquid chromatography
It was essential in the method development for parameters opti-
mization such as selection of types of reversed-phase chromato-
graphic column and organic solvent, composition of the mobile
phase, flow rate, etc. Reversed-phase chromatographic column
containing SpursilTM C18 5 mm column (150  2.1 mm; Dikma,
USA), Gemini 5 m C18 110A (50  2 mm 5 mm; Phenomenex,
USA) and Analytical ODS 5 mm (4.6  150 mm; Agilent, USA)
were tested to improve peak shape. Finally, a Spursil TM C18
5 mm column (150  2.1 mm; Dikma, USA) was suitable for the
chromatographic separation because under the current LC con-
ditions, the column provided excellent results in terms of re-
sponse, retention time and peak shapes. The mobile phase
Table II
Matrix Effect of LA in Rat Plasma (n ¼ 5)
Route of LA concentration MC (mean + SD)
administration (ng/mL) (ng/mL)
i.g. 10 10.26 + 0.65
500 467.54 + 10.75
1,000 989.76 + 45.53
i.v. 25 23.45 + 0.75
1,000 943.12 + 42.44
3,000 2879.43 + 37.43
MC, measured concentration.
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
flow rate of 0.25 mL/min, because methanol and ammonium ac-
etate were able to get the lowest background noise and the best
resolution. The chromatographic run time for one injection was
5 min. The method exhibited the baseline separation of LA and IS
were approximately at 2.0 and 3.0 min with a good peak shape
and symmetry, respectively (Figure 2). Figure 2 shows represen-
tative LC – MS-MS chromatograms of blank plasma, LLOQ of LA
and IS spiked with blank plasma (5 and 10 ng/mL, respectively),
blank plasma spiked with 70 ng/mL of LA and 150 ng/mL of IS
and rat plasma at 1.5 h following a single oral administration of
DSS at 20 mg/kg.
Mass spectrometry
In our study, to develop an optimal MS-MS method for detection
of LA in biological samples, a comparison between the ionization
of LA under the ESI negative and positive ion modes was carried
out. Our research results showed that a better response with
good sensitivity, reproducibility and fragmentation as well as
stronger intensity of the most abundant molecule ion for LA
was produced in the positive ion mode than that in the negative
mode. Under the same MS detector conditions, AD (IS), as a re-
sult of its similarity in chemical structure to LA, could be ionized
efficiently as well. Thus, the positive ion mode was chosen for
analysis of LA throughout this study. Furthermore, precursor
ions were fragmented and unique product ions were measured
in MRM, which enabled selective detection of all compounds
simultaneously. The sensitive ions in the full-scan positive ion
mass spectra were 398.5 [MþNH4]þ ! 381.3 [MþNH4 – OH]þ
for LA and 368.0 [MþNH4]þ ! 351.1 [MþNH4 – OH]þ for AD
(IS). The transitions of m/z 398.5 precursor ion to the m/z
381.3 were used for quantification for LA. Similarly, for AD m/z
368.0 precursor ion to the m/z 351.1 was also used for the quan-
Precision (%) Accuracy (%) tification of the analytes.
6.30 2.53 Validation procedures
2.30 6.94
4.60 1.03
3.20 6.61 Matrix effect
4.50 6.03 In this work, the matrix effects of plasma samples were evaluated
1.30 4.19
by determining QC samples under the optimized assay condi-
tions. The results are presented in Table II, all of them were

Table III
Stability Tests of LA in Plasma (n ¼ 5)

Route of LA concentration Three freeze – thaw cycles Short-term stability for 24 h in plasma Long-term stability for 30 days at 2708C
administration (ng/mL) (mean + SD)
MC (mean + SD) Precision Accuracy MC (mean + SD) Precision Accuracy MC (mean + SD) Precision Accuracy
(ng/mL) (%) (%) (ng/mL) (%) (%) (ng/mL) (%) (%)
i.g. 10 9.97 + 0.47 4.7 0.30 9.63 + 0.41 4.20 3.84 9.23 + 0.43 4.66 8.34
500 494.32 + 12.85 2.6 1.15 472.94 + 23.54 4.98 5.72 496.12 + 26.05 5.25 0.78
1,000 938.47 + 54.43 5.8 6.56 984.21 + 32.46 3.30 1.60 965.45 + 47.59 4.93 3.58
i.v. 25 24.14 + 1.04 4.3 3.56 24.31 + 1.12 4.61 2.84 23.84 + 1.31 5.49 4.87
1,000 973.91 + 13.63 1.4 2.68 964.31 + 36.12 3.75 3.70 998.46 + 36.72 3.68 0.15
3,000 2901.34 + 69.63 2.4 3.40 2846.78 + 50.46 1.77 5.38 2798.46 + 46.94 1.68 7.20

MC, measured concentration.

LC– MS-MS determination of LA 1269

within the acceptable limit, which indicated that no significant range. The calibration curve for oral administration was prepared
ion suppression or enhancement from the plasma matrix was ob- by a typical equation y ¼ 0.0024x þ 0.0037, and the calibration
served on the detection. curve for i.v. administration was obtained by a typical equation
y ¼ 0.002x þ 0.005, the best fit of peak area ratios of LA to IS
Stability (y) versus the plasma concentration of LA (x, ng/mL). The aver-

The stability experiments were aimed at testing the possible con-
ditions to which the samples might be exposed during storage
and handling. The post-preparative stability indicated that LA
was stable at room temperature for 24 h. LA could also withstand
three cycles of the freeze – thaw process for QC samples. QC
samples were stable when stored frozen at 2208C for at least
30 days. All results of the stability tests are summarized in
Table III, indicated no appreciable degradation occurred for
routine analysis.
Extraction recovery
The extraction recovery was determined in six replicates by
comparing the peak areas of the extracted plasma at three QC
levels with those obtained from the direct injection of standard
solutions without preparation at the same concentrations.
Liquid/liquid extraction has been commonly applied in biologi-
cal fluids by different solvents. Therefore, we tested the extrac-
tion recoveries of LA using different organic solvents including
acetonitrile, methanol and their mixtures. But we observed bad
peak shapes after that. Subsequently, SPE was investigated as sam-
ples pre-treatment technique. Considering their lipophilic char-
acterization, the extracted recovery with Dikma C18-SPE column
was high. Table IV summarizes the results.
Calibration curve, linearity and LLOQ
The calibration curves were established by eight calibration stan-
dards for oral administration and intravenous administration,
respectively. The calibration standard curve had a reliable repro-
ducibility over the standard concentrations across the calibration
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
age correlation coefficient (r) was .0.9993, exhibited to be lin-
ear over the calibration range for LA. The LLOQ samples of six
different rodents’ plasma was defined as the LLOQ in the calibra-
tion curve that can be accurately measured. The LLOQ for LA
were analyzed and found to be 5.0 ng/mL, with an accuracy of
108.8% and precision of 8.4%. The LLOQ for IS was selected to
be 10.0 ng/mL, with an accuracy of 104.2% and precision of
9.2%, respectively. The LLOQ was based on S/N 10, where
the limit is sufficient for rat pharmacokinetic studies following
oral administration and i.v. administration of LA.
Accuracy and precision
The intra-day accuracy and precision in rat plasma were evaluat-
ed using the low, medium and high QC (n ¼ 5) spiked plasma
samples. One reproduction of the QC samples at each concentra-
tion level from three separate validation batches was used to
determine the inter-day precision. As summarized in Table V,
the obtained results suggested that the method was judged to
have a satisfactory accuracy and precision.
Selectivity and specificity
The selectivity and specificity were assessed by comparing the
typical chromatographic profiles of five different batches of drug-
free rat plasma with the corresponding spiked plasma samples
spiked with LA and the IS. The retention times of LA and the IS
were 2.0 and 3.0 min. No endogenous substances were ob-
served to interfere with LA and the IS under the specified LC –
MS-MS conditions.

Table IV
Recovery of LA in Rat Plasma by the Dikma C18-SPE Method (n ¼ 5)
Pharmacokinetic study
Route of LA concentration (ng/ MC (mean + SD) RSD (%) Recovery (%) The newly validated method was successfully applied to the
administration mL) (ng/mL)
pharmacokinetics of LA in rat plasma after i.v. and i.g. adminis-
i.g. 10 9.65 + 0.50 5.18 96.5 tration at a dose of 2 and 20 mg/kg, respectively. Furthermore,
500 478.24 + 37.30 7.80 95.6
1,000 975.42 + 27.31 2.80 97.5 the mean plasma concentration – time profile of LA after oral
i.v. 25 26.13 + 1.67 6.39 104.5 administration and intravenous administration to rats is
1,000 962.41 þ 40.42 4.20 96.2
3,000 2973.35 þ 371.67 12.50 99.1
shown in Figure 3. The major pharmacokinetic parameters
were calculated by a non-compartmental model and are listed
MC, measured concentration. in Table VI.

Table V
Intra- and Inter-Day Precision and Accuracy for Analysis of LA in Plasma

Route of administration LA concentration (ng/mL) Intra-day (n ¼ 5) Inter-day (n ¼ 5)

MC (mean + SD) (ng/mL) Precision (%) Accuracy (%) MC (mean + SD) (ng/mL) Precision (%) Accuracy (%)
i.g. 10 9.76 + 1.33 13.63 2.46 9.23 + 0.25 2.7 8.34
500 484.51 + 12.62 2.60 3.20 475.62 + 8.56 1.8 5.13
1,000 963.25 + 27.93 2.90 3.82 981.64 + 18.65 1.9 1.87
i.v. 25 24.69 + 0.91 3.69 1.26 24.01 + 1.12 4.6 4.12
1,000 937.17 + 47.86 5.11 6.70 941.27 + 21.65 2.3 6.24
3,000 2976.64 + 116.09 3.90 0.79 2802.34 + 182.15 6.5 7.05

MC, measured concentration.

1270 He et al.
Discussion
The statistical analysis between three oral groups was performed
using ANOVA. Compared with the LA group and the Rhizoma
Chuanxiong group, the AUC0 ! 1 values of the DSS group
were remarkably improved (P , 0.05). Furthermore, the Tmax
ipated in vitro metabolism of ligustilide, which is one kind of es-
sential ingredients belonging to Rhizoma Chuanxiong and A.
sinensis (14, 23). In addition, it has been verified that CYP3A
plays a primary role in tetramethylpyrazine metabolism in rats
(28). The ingredients like ligustilide and tetramethylpyrazine in

of the DSS group was longer than that of the LA group and the
Rhizoma Chuanxiong group. The higher AUC0! 1 values and
the longer Tmax always mean that the analytes could stay longer
and be absorbed better in the body, leading to alerting better or
longer efficacy. That means the DSS may have better efficacy than
Rhizoma Chuanxiong or LA. There may be certain ingredients
in DSS which have an effect on improving the absorption of LA,
which may prolong its residence time and get the large AUC0! 1
values, and thus prolong its efficacy.
It is also commonly reported for the pharmacokinetics of tra-
ditional Chinese medicine (TCM) that the coexisting constitu-
ents in herbal extracts may serve on substrates of inducers or
inhibitors of cytochrome enzymes (27). It has been proven that
CYP3A4, CYP2C9 and CYP1A2 are the major isoenzymes partic-
Figure 3. Mean plasma concentration– time profile of LA (filled diamonds), Rhizoma
Chuanxiong ( filled squares), DSS ( filled triangles) in rat plasma after oral
administration to rats (n ¼ 8) (A), and the mean plasma concentration – time profile
of LA (filled circles) in rat plasma after intravenous administration to rats (n ¼ 8) (B).
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
DSS may act as substrates of inhibitors or inducers of cytochrome
enzymes and thus they may effect on the pharmacokinetics of LA
(29). However, further research studies are required to verify
these relationships.
The oral bioavailability calculated by (AUC i.g./dose)/(AUC
i.v./dose) of LA in rats was 7.5%. It was reported that an army
of effective ingredients in herbs that possess low bioavailability
(30) may be attributed to extensive first-pass effect or the solu-
bility of drugs (31– 33), or even to be affected by coexisting con-
stituents in the herbal extracts. For example, it was also proved
that P. lactiflora Pall could reduce the plasma concentration
and bioavailability of ligustilide, the dimer of LA; however, the
P. cocos – A. macrocephala – Alismatis group contributes to in-
crease its bioavailability (34). More studies to elucidate this hy-
pothesis are warranted.
Because DSS is used in blood-activating and stasis-eliminating
drugs, it is possible that the increase in ligustilide elimination
(P , 0.05) may be an immediate result of hepatic blood flow
change (35). Meanwhile, A. sinensis (36) of DSS could be respon-
sible for activating blood circulation and R. Chuanxiong is
proved to have effect on the influence for the blood velocity
(32, 37), so that they could help to accelerate the elimination
of LA. Nevertheless, to date there is no certain evidence to sup-
port this contention. Moreover, in most of the cases, clearance of
LA and ligustilide was significantly higher than that of the hepatic
blood flow in rats (38), implying that in addition to widespread
metabolism, other factors might also contribute to the fast elim-
ination of LA and ligustilide from systemic circulation.
This study demonstrates that important information for inves-
tigating the pharmacokinetics and oral bioavailability of LA,
Rhizoma Chuanxiong and DSS extract in vivo. The results
were complex and require more detailed research works.
Further studies concerning the relationship among the pharma-
cokinetics of LA and other ingredients of DSS should be per-
formed to obtain more effectual information for clinical
practice after the administration of DSS. The results of such stud-
ies should contribute to the feasibility of the effective therapeu-
tic usage of TCM (39).

Table VI
Major Pharmacokinetic Parameters of LA after Oral Administration and Intravenous Administration (n ¼ 8)

Analyte Parameters (mean + SD)

kb (h21) ka (h21) ka (h21) t1/2b (h) t1/2a (h) t1/2a (h)
LA (i.g.) 0.183 + 0.02 0.781 + 0.08 0.815 + 0.06 3.789 + 0.38 0.857 + 0.09 0.887 + 0.08
DSS (i.g.) 0.250 + 0.01a 1.871 + 0.15a 1.860 + 0.17a 2.767 + 0.23a 0.37 + 0.02a 0.372 + 0.03a
Chuanxiong (i.g.) 0.217 + 0.03 0.359 + 0.04a 0.407 + 0.03a 3.192 + 0.29 1.929 + 0.14a 1.701 + 0.17a
LA (i.v.) 0.289 + 0.02a 0.266 + 0.01a 2.402 + 0.15a 0.265 + 0.03a 0.266 + 0.18a
Tmax (h) Cmax (ng/mL) AUC0! t (ng h/mL) AUC0! 1 (ng h/mL)
LA (i.g.) 0.5 + 0.05 1149.2 + 103.43 3816.1 + 343.44 3848.2 + 307.84
DSS (i.g.) 2.0 + 0.10a 1948.6 + 136.65a 11198.1 + 1049.85a 11211.7 + 380.20a
Chuanxiong (i.g.) 0.75 + 0.06 1173.6 + 117.36 5203.1 + 468.28 5247.9 + 419.84
LA (i.v.) 3988.5 + 319.04a 50691.3 + 1032.68a 50746.5 + 2613.62a

kb, rate of elimination, ka, rate of absorption, ka, rate of distribution.

a
P , 0.05 versus the LA (i.g.) group.

LC– MS-MS determination of LA 1271

Conclusions
In this study, a rapid, simple, reproducible and suitable LC –
MS-MS method was developed and validated for the quantifica-
tion of LA with AD as IS in rat plasma with high recovery and
minimal matrix effect. The assay utilized a SPE as sample clean-up
procedure and a reversed-phase separation with sufficient selec-
tivity and sensitivity. The results suggest that the method is able
to be used in the drug metabolism and pharmacokinetic study of
LA and other Chinese medicinal preparations containing LA,
which is worthy of further studies. The accurate and precise
method has been successfully applied to compare the pharmaco-
kinetic behaviors between oral and intravenous groups, and the
oral bioavailability was calculated, which can be a reference of
pharmacology research related to Rhizoma Chuanxiong. DSS
is convenient to be dosed and with low risks. In the meanwhile,
it not only has a favorable absorption after oral intake, but also
lasts a longer efficacy.
Acknowledgments
This research was supported by the National Natural Science
Foundation of China (no. 81273817), Doctoral Fund of Education
Ministry of China (no. 20114425110007), Guangdong Provincial
Major Science and Technology for Special Program of China (no.
2012A080202017), Guangdong Provincial Department of Science
and Technology Foundation of China (no. 2010A030100009),
Guangdong Provincial Natural Science Foundation of China
(no. S2012040006514), the Scientific and Technical Innovation
Project of Guangdong Provincial Education Department of
China (2012KJCX0032) and the Characteristic Key Discipline
Construction Fund of Chinese Internal Medicine of Guangzhou
University of Chinese Medicine.
References
1. Huang, Z., Mao, Q.Q., Zhong, X.M., Li, Z.Y., Qiu, F.M., Ip, S.P.; Mechanistic
study on the antidepressant-like effect of Danggui-Shaoyao-San, a
Chinese herbal formula; Evidence-Based Complementary and
Alternative Medicine, (2012); 2012: 173565.
2. Pigeon, W.R.; Insomnia as a predictor of depression: do insomnia sub-
types matter?; Sleep, (2010); 33: 1693– 1702.
3. Chen, L.L., Qi, J., Chang, Y.X., Zhu, D.N., Yu, B.Y.; Identification and
determination of the major constituents in traditional Chinese me-
dicinal formula Danggui-Shaoyao-San by HPLC – DAD – ESI-MS/MS;
Journal of Pharmaceutical and Biomedical Analysis, (2009); 50:
127–137.
4. Kou, J., Zhu, D., Yan, Y.; Neuroprotective effects of the aqueous ex-
tract of the Chinese medicine Danggui-Shaoyao-San on aged mice;
Journal of Ethnopharmacology, (2005); 97: 313– 318.
5. Hu, Z.Y., Liu, G., Yuan, H., Yang, S., Zhou, W.X., Zhang, Y.X., et al.
Danggui-Shaoyao-San and its active fraction JD-30 improve
Ab-induced spatial recognition deficits in mice; Journal of
Ethnopharmacology, (2010); 128: 365– 372.
6. Huang, Y., Hu, Z.Y., Yuan, H., Shu, L., Liu, G., Qiao, S.Y., et al.;
Danggui-Shaoyao-San improves learning and memory in female
SAMP8 via modulation of estradiol; Evidence-Based Complementary
and Alternative Medicine, (2014); 127: 3630–3637.
7. Yu, Y.J., Song, X.Y., Cai, J.Z., Yang, X.J., Yan, Y.L.; Effects of Danggui
Shaoyao San on experimental myocardial ischemia induced by pitui-
tary injection in mice; Chinese Journal of Experimental Traditional
Medical Formulae, (2012); 18: 243–246.
8. Chinese Pharmacopoeia Committee; Chinese pharmacopoeia, Part
1. Chemical Industry Press, Beijing, China, (2010), 38 pp.
1272 He et al.
9. Li, W.X., Tang, Y.P., Chen, Y.Y., Duan, J.A.; Advances in the chemical
analysis and biological activities of chuanxiong; Molecules, (2012);
17: 10614 –10651.
10. Shuai, L., Chen, Z., Gao, R.X., Yang, T.M.; Interaction of lobed kudzu-
vine root, rhizoma chuanxiong with both acetylcholinesterase and
beta-amyloid(Ab1242); Pharmacognosy Magazine, (2013); 9:
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
196– 201.
11. Lin, Y.L., Lee, Y.C., Huang, C.L., Lai, W.L., Lin, Y.R., Huang, N.K.;
Ligusticum chuanxiong prevents rat pheochromocytoma cells
from serum deprivation-induced; Journal of Ethnopharmacology,
(2007); 109: 428–434.
12. Lee, T.F., Lin, Y.L., Huang, Y.T.; Studies on antiproliferative effects of
phthalides from Ligusticum chuanxiong in hepatic stellate cells;
Planta Medica, (2007); 73: 527–534.
13. Tang, Y., Zhu, M., Yu, S., Hua, Y., Duan, J.A., Su, S., et al.; Identification
and comparative quantification of bio-active phthalides in essential
oils from si-wu-tang, fo-shou-san, radix angelica and rhizoma chuan-
xiong; Molecules, (2010); 15: 341– 351.
14. Li, S.L., Yan, R., Tam, Y.K., Lin, G.; Post-harvest alteration of the main
chemical ingredients in Ligusticum chuanxiong HORT. (Rhizoma
Chuanxiong); Chemical & Pharmaceutical Bulletin, (2007); 55:
140– 144.
15. Zuo, A., Wang, L., Xiao, H., Li, L., Liu, Y., Yi, J.; Identification of the ab-
sorbed components and metabolites in rat plasma after oral adminis-
tration of Rhizoma Chuanxiong decoction by HPLC-ESI-MS/MS;
Journal of Pharmaceutical and Biomedical Analysis, (2011); 56:
1046–1056.
16. Guzman, J.D., Evangelopoulos, D., Gupta, A., Prieto, J.M., Gibbons, S.,
Bhakta, S.; Antimycobacterials from lovage root (Ligusticum offici-
nale Koch); Phytotherapy Research, (2013); 27: 993–998.
17. Chen, F., Wang, T., Wang, J., Wang, Z.Q., Qian, M.; Levistolide A over-
comes P-glycoprotein-mediated drug resistance in human breast car-
cinoma cells; Acta Pharmacologica Sinica, (2008); 29: 458– 464.
18. Wang, J.; Study on synergic anticancer effect of levistolide A and
vinorelbine on human non-small cell lung carcinoma. School of
Life Science of East China Normal University, China, (2007), pp.
19 –45.
19. Yi, T., Leung, K.S., Lu, G.H., Chan, K., Zhang, H.; Simultaneous qualita-
tive and quantitative analyses of the major constituents in the rhi-
zome of Ligusticum chuanxiong using HPLC-DAD-MS; Chemical
& Pharmaceutical Bulletin, (2006); 54: 255–259.
20. U.S. Food and Drug Administration (FDA); Guidance for industry:
bioanalytical methods validation, (2001).
21. US Department of Health and Human Services Food and Drug
Administration, Center for Drug Evaluation and Research (CDER),
Guidance for industry, bioanalytical method validation (2001),
Center for Veterinary Medicine, (2001).
22. Li, Y.Y., Song, Y.Y., Liu, C.H., Huang, X.T., Zheng, X., Li, N., et al.;
Simultaneous determination of esculin and its metabolite esculetin
in rat plasma by LC–ESI-MS/MS and its application in pharmacokinet-
ic study; Journal of Chromatography B, (2012); 907:27– 33.
23. Matuszewski, B., Constanzer, M., Chavez-Eng, C.; Strategies for the as-
sessment of matrix effect in quantitative bioanalytical methods based
on HPLC-MS/MS; Analytical Chemistry, (2003); 75: 3019.
24. Zheng, X., Huang, X.T., Li, N., Li, Y.Y., Mi, S.Q., Wang, N.S., et al.;
Determination of geniposidic acid in rat plasma by LC –MS/MS and
its application to in vivo pharmacokinetic studies; Journal of
Chromatography B, (2012); 887–888: 138– 142.
25. Pei, L., Bao, Y., Ma, L., Wang, Q., Ye, Y., Han, X., et al.; A sensitive meth-
od for determination of platycodin D in rat plasma using liquid chro-
matography/tandem mass spectrometry and its application to a
pharmacokinetic study; Planta Medica, (2012); 78: 244– 251.
26. Chiou, W.L.; Critical evaluation of the potential error in pharmacoki-
netic studies of using the linear trapezoidal rule method for the cal-
culation of the area under the plasma level –time curve; Journal of
Pharmacokinetics and Biopharmaceutics, (1978); 6: 539.
27. Zhang, L., Yan, R., Su, R., Yang, C., Liu, S., Yu, X., et al.; Bioavailability
enhancement of osthole after oral administration of Bushen Yizhi
prescription extract to rats followed by Cnidium monnieri (L.)
 Cusson fruits extract in comparison to pure osthole at different
doses; Journal of Ethnopharmacology, (2014); 152: 266– 271.
28. Kuang, X.D., Li, X.H., Xiong, Y.Q.; Study on metabolism of tetrame-
thylpyrazine in system of rat liver microsomes; Zhongguo Zhong
Yao Za Zhi, (2006); 31: 1971–1975.
29. Qian, M., Shi, L.F., Hu, J.H.; Enzyme kinetics of ligustilide metabolism
in rat liver microsomes; Yao Xue Xue Bao, (2009); 44: 395–400.
30. Yang, K.Y., Lin, L.C., Tseng, T.Y., Wang, S.C., Tsai, T.H.; Oral bioavail-
ability of curcumin in rat and the herbal analysis from Curcuma
longa by LC-MS/MS; Journal of Chromatography B, (2007); 853:
183–189.
31. Zhang, X.Y., Qiao, H., Shi, Y.B.; HPLC method with fluorescence detec-
tion for the determination of ligustilide in rat plasma and its pharmaco-
kinetics; Journal of Pharmacokinetics and Biopharmaceutics,
(2014); 52: 21–30.
32. Wong, S.M., Kellaway, I. W., Murdan, S.; Enhancement of the dissolu-
tion rate and oral absorption of a poorly water soluble drug by forma-
tion of surfactant containing microparticles; International Journal
of Pharmaceutics, (2006); 317: 61– 68.
33. Patravale, V. B., Date, A.A., Kulkami, R.; Nanosuspensions: a promising
drug delivery strategy; Journal of Pharmacy and Pharmacology,
(2004); 56: 827–840.
34. Chen, L.L., Qi, J., Kou, J.P., Yu, B.Y., Zhu, D.N.; Effects of compatibility
of Danggui-Shaoyao-San on pharmacokinetics of major components
in rats; Chinese Journal of Experimental Traditional Medical
Formula, (2012); 18: 121–124.
35. Yan, R., Ko, N.L., Li, S.L., Tam, Y.K., Lin, G.; Pharmacokinetics and me-
tabolism of ligustilide, a major bioactive component in Rhizoma
Chuanxiong, in the rat; Drug Metabolism and Disposition,
(2008); 36: 400– 408.
Downloaded from https://academic.oup.com/chromsci/article-abstract/53/8/1265/312117 by Universidad Nacional Autonoma de Mexico user on 17 January 2020
36. Li, W., Guo, J., Tang, Y., Wang, H., Huang, M., Qian, D., et al.;
Pharmacokinetic comparison of ferulic acid in normal and blood defi-
ciency rats after oral administration of Angelica sinensis, Ligusticum
chuanxiong and their combination; International Journal of
Molecular Sciences, (2012); 13: 3583–3597.
37. Liu, J.J., Zheng, S.L., Fan, Q.J., Yuan, J.C., Yang, S.M., Fan-Lei Kong, F.L.;
Optimization of high-pressure ultrasonic-assisted simultaneous ex-
traction of six major constituents from Ligusticum chuanxiong
Rhizome using response surface methodology; Molecules, (2014);
19: 1887–1911.
38. Davies, B., Morris, T.; Physiological parameters in laboratory animals
and humans; Phytotherapy Research, (1993); 10: 1093– 1095.
39. Zeng, M., Zhang, J., Yang, Y., Jin, Y., Xiao, W., Wang, Z., et al.; An au-
tomated dual-gradient liquid chromatography – MS/MS method for
the simultaneous determination of ferulic acid, ligustrazine and ligu-
stilide in rat plasma and its application to a pharmacokinetic study;
Journal of Pharmaceutical and Biomedical Analysis, (2014); 88:
354– 363.
LC– MS-MS determination of LA 1273